Subbarao Kambhampati (కంభంపాటి సుబ్బారావు)
@rao2z.bsky.social
AI researcher & teacher at SCAI, ASU. Former President of AAAI & Chair of AAAS Sec T. Here to tweach #AI. YouTube Ch: http://bit.ly/38twrAV Twitter: rao2z
📣 New in the AAAI AI Magazine (open access link 👉https://onlinelibrary.wiley.com/doi/epdf/10.1002/aaai.70113 )
I am rather pleased with how this talk reflects my current thinking on a bunch of LLM reasoning related issues. www.youtube.com/watch?v=mD32...
On the Mythos of LLM "Thinking Traces" (Talk @ Tubingen Ellis/MPI 10/9/2026)
YouTube video by Subbarao Kambhampati
youtube.com
🐜🐜🐜ANThropomorphism..🐜🐜🐜 x.com/rao2z/status...
Subbarao Kambhampati (కంభంపాటి సుబ్బారావు) (@rao2z) on X
In a recent harrowing development, a very normal routine experiment by the frontier company Ant went completely off the rails. 😱 Ant put thousands of their ant agents in a pretty secure ant farm with...
x.com
I recommend this article about AI reasoning, where the author lets us in on his struggles w/ AI cognitive dissonance. Plus some priceless quotes from @rao2z.bsky.social. (My recommendation has *nothing* to do with the fact that I'm quoted in it too 😇) www.quantamagazine.org/is-ai-reason...
Is AI Reasoning Right for the Wrong Reasons? | Quanta Magazine
The idea that artificial intelligence can “reason” is more intuitive than ever. But intuitions can be wrong, and the science is far from settled.
quantamagazine.org
🥳 📢 Our Beyond Semantics paper on LLM "thinking traces" is now officially published at @tmlrorg.bsky.social (with a J2C certification--so you will also see us at #NeurIPS2026..😋) TMLR 👉 openreview.net/forum?id=gDE... Kudos to Karthik Valmeekam, Vardhan Palod, Kaya Stechly and Atharva Gundawar!
Thoughts on Mechanistic Interpretability breakthroughs.. www.linkedin.com/posts/subbar...
#aiaphorisms | Subbarao Kambhampati
Mechanistic Interpretability research be like.. We wanted to see what the bread slice is thinking when it is in the toaster. Is it taking its singing equanimously? thinking religious thoughts? or plo...
linkedin.com
Here is the recording of a keynote talk I gave at #ICLR2026 UCRL workshop a few days back on the challenges of explainable AI in the era of LLMs. It looks at how LLMs seemingly make some XAI problems go away, while bringing in a new set.. www.youtube.com/watch?v=sM-0...
What does XAI mean in the era of LLMs? (Keynote @ ICLR 2026 Unified Concept Representations wkshp)
YouTube video by Subbarao Kambhampati
youtube.com
Thinking Trace Trilogy: Here is our trilogy of refereed papers investigating the interpretability of the thinking traces output by reasoning models: arxiv.org/abs/2504.09762 in #ICML2026; arxiv.org/abs/2505.13792 in #ACL2026 and arxiv.org/abs/2505.13775 in @tmlrorg.bsky.social 2026
Speaking in a couple of hours at UCRL workshop #ICLR2026 (Sunday 2:15pm Rio time; Room 209; iclr.cc/virtual/2026... )
Looking forward to speak at the #ICLR2026 @iclr-conf.bsky.social workshop on "Unifying Concept Representation Learning" this Sunday (Rm 209; 2:15PM Rio time).. with a title inspired by Bernard Shaw's quote about England and America.. 😛 1/2
𝙈𝙞𝙨𝙘𝙤𝙣𝙘𝙚𝙥𝙩𝙞𝙤𝙣𝙨 𝙖𝙗𝙤𝙪𝙩 𝙇𝙇𝙈-𝙈𝙤𝙙𝙪𝙡𝙤 𝙖𝙣𝙙 𝙩𝙝𝙚 𝙍𝙤𝙡𝙚 𝙤𝙛 𝙑𝙚𝙧𝙞𝙛𝙞𝙚𝙧𝙨 𝙞𝙣 𝙍𝙚𝙖𝙨𝙤𝙣𝙞𝙣𝙜 𝙈𝙤𝙙𝙚𝙡𝙨. #SundayHarangue www.linkedin.com/pulse/miscon...
Misconceptions about LLM-Modulo and the Role of Verifiers in Reasoning Models. #SundayHarangue
I have talked before about LLM-Modulo framework (c.f.
linkedin.com
Our paper questioning the wide-spread anthropomorphization of LRM intermediate tokens as "reasoning traces" has just been accepted to @tmlrorg.bsky.social (arxiv.org/abs/2505.13775). This work was lead by the dream team of Karthik Valmeekam, Vardhan Palod, Kaya Stechly and Atharva Gundawar. 1/
Verifying LLM problem progress with LLM-Process-Modulo www.linkedin.com/posts/subbar...
𝐋𝐋𝐌-𝐏𝐫𝐨𝐜𝐞𝐬𝐬-𝐌𝐨𝐝𝐮𝐥𝐨: Our original 𝘓𝘓𝘔-𝘔𝘰𝘥𝘶𝘭𝘰 framework (https://lnkd.in/gyAyKx4E) is a Generate-Test framework, with the LLM generating candidate solutions and a bank of… | Subbarao Kambhampati
𝐋𝐋𝐌-𝐏𝐫𝐨𝐜𝐞𝐬𝐬-𝐌𝐨𝐝𝐮𝐥𝐨: Our original 𝘓𝘓𝘔-𝘔𝘰𝘥𝘶𝘭𝘰 framework (https://lnkd.in/gyAyKx4E) is a Generate-Test framework, with the LLM generating candidate solutions and a bank of verifiers critiquing those sol...
linkedin.com
World Models: The old, the new and the wishful #SundayHarangue There is a lot of chatter about world models of late--even more than can be explained by Yann's bet. I was going to comment on this clamor in my next class, and thought I will preview it here first..😋 www.linkedin.com/pulse/world-...
World Models: The old, the new and the wishful #SundayHarangue
There is a lot of chatter about world models of late--even more than can be explained by Yann betting his entire new enterprise on it. I was going to comment on this clamor in my class this week, and ...
linkedin.com
If you, as a CS Prof, are wondering whether you need PhD students at all now that you have wangled a subscription to Claude Code, your lab probably had a pretty depressing vibe to begin with--and 'em students are likely better off with you hanging out with Claude.. #AIAphorisms
Our initial interest in the reasoning capabilities of LLMs arose from our proximal work on Human-AI interaction. It was thus gratifying to give a keynote on the role of LLMs in Human-AI interaction at IEEE CogSIMA conference yesterday Video 👉 youtu.be/yf4RQYlKRJI
Role of LLMs in Human-AI Interaction (Keynote @ Cogsima 2026)
YouTube video by Subbarao Kambhampati
youtu.be
We updated our position on anthropomorphization of intermediate tokens in LRMs--with additional results and a call to action.. arxiv.org/abs/2504.09762
𝗣𝗼𝘀𝘁-𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗼𝗿 𝘄𝗶𝘁𝗵𝗼𝘂𝘁 𝘀𝗲𝗹𝗳-𝗱𝗶𝘀𝘁𝗶𝗹𝗹𝗮𝘁𝗶𝗼𝗻 𝗶𝘀 𝗮𝗹𝗹 𝗮𝗯𝗼𝘂𝘁 𝗰𝗼𝗺𝗽𝗶𝗹𝗶𝗻𝗴 𝘃𝗲𝗿𝗶𝗳𝗶𝗲𝗿 𝘀𝗶𝗴𝗻𝗮𝗹 𝗶𝗻𝘁𝗼 𝘁𝗵𝗲 𝗯𝗮𝘀𝗲 𝗟𝗟𝗠 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗼𝗿 #SundayHarangue 𝗦𝘁𝗮𝗻𝗱𝗮𝗿𝗱 𝗣𝗼𝘀𝘁-𝗧𝗿𝗮𝗶𝗻𝗶𝗻𝗴 ≈ 𝗖𝗼𝗺𝗽𝗶𝗹𝗶𝗻𝗴 𝗩𝗲𝗿𝗶𝗳𝗶𝗲𝗿 𝗦𝗶𝗴𝗻𝗮𝗹 𝗦𝗲𝗹𝗳-𝗱𝗶𝘀𝘁𝗶𝗹𝗹𝗮𝘁𝗶𝗼𝗻 = 𝗖𝗼𝗺𝗽𝗶𝗹𝗶𝗻𝗴 𝗩𝗲𝗿𝗶𝗳𝗶𝗲𝗿 𝗙𝗲𝗲𝗱𝗯𝗮𝗰𝗸 𝘁𝗼𝗼 See 👇 for more.. www.linkedin.com/posts/subbar...
Planning & Reasoning Abilities of LLMs/LRMs (Lecture 2 @ Melbourne ML Summer School 2026) | Subbarao Kambhampati
𝗣𝗼𝘀𝘁-𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗼𝗿 𝘄𝗶𝘁𝗵𝗼𝘂𝘁 𝘀𝗲𝗹𝗳-𝗱𝗶𝘀𝘁𝗶𝗹𝗹𝗮𝘁𝗶𝗼𝗻 𝗶𝘀 𝗮𝗹𝗹 𝗮𝗯𝗼𝘂𝘁 𝗰𝗼𝗺𝗽𝗶𝗹𝗶𝗻𝗴 𝘃𝗲𝗿𝗶𝗳𝗶𝗲𝗿 𝘀𝗶𝗴𝗻𝗮𝗹 𝗶𝗻𝘁𝗼 𝘁𝗵𝗲 𝗯𝗮𝘀𝗲 𝗟𝗟𝗠 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗼𝗿 #SundayHarangue 𝗦𝘁𝗮𝗻𝗱𝗮𝗿𝗱 𝗣𝗼𝘀𝘁-𝗧𝗿𝗮𝗶𝗻𝗶𝗻𝗴 ≈ 𝗖𝗼𝗺𝗽𝗶𝗹𝗶𝗻𝗴 𝗩𝗲𝗿𝗶𝗳𝗶𝗲𝗿 𝗦𝗶𝗴𝗻𝗮𝗹 𝗦𝗲𝗹𝗳-𝗱𝗶𝘀𝘁𝗶𝗹𝗹𝗮𝘁𝗶𝗼...
linkedin.com
Does it really make sense to think of inference efficiency in terms of the number of tokens produced? No. 👇 x.com/i/status/202...
x.com
Here are the recordings of two lectures on 𝗣𝗹𝗮𝗻𝗻𝗶𝗻𝗴 & 𝗥𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴 𝗖𝗮𝗽𝗮𝗯𝗶𝗹𝗶𝘁𝗶𝗲𝘀 𝗼𝗳 𝗟𝗟𝗠𝘀/𝗟𝗥𝗠𝘀 that I gave this week at Melbourne ML Summer School (lnkd.in/g7rxg9sw). 𝙇𝙚𝙘𝙩𝙪𝙧𝙚 1: youtube.com/watch?v=_PPV... 𝙇𝙚𝙘𝙩𝙪𝙧𝙚 2: youtube.com/watch?v=fKlm...
A common theme in our work these past few years has been pushing back on facile anthropomorphizations (and/or efforts that bring questionable/discredited Cognitive Science metaphors) to LLMs.. So I enjoyed giving this talk at @ivado.bsky.social yesterday... www.youtube.com/watch?v=CoyS...
Anthropomorphization Sins in Modern AI (or Perils of Prematurely Applying Lens of Cognition to LLMs)
YouTube video by Subbarao Kambhampati
youtube.com
Three of my talks in India last month--at @iitdelhi.bsky.social, @msftresearch.bsky.social India and at IndoML Symposium--were "On the Mythos of LRM Thinking Tokens." Here is a recording of one of them--the talk I gave at MSR India. www.youtube.com/watch?v=fCQX...
On the Mythos of LRM "Thinking Tokens" (Talk @ Microsoft Research, India; 12/16/2025)
YouTube video by Subbarao Kambhampati
youtube.com
ICYMI, here is my keynote on the semantics of LRM "thinking traces" at #NeurIPS2025 workshop on Multimodal Algorithmic Reasoning. It's a unified view of the seven papers we presented at the conference workshops. Special thanks to the engaged audience..🙏 www.youtube.com/watch?v=rvby...
Talk on the semantics of "Thinking Traces" (Keynote at NeurIPS2025 MAR Workshop)
YouTube video by Subbarao Kambhampati
youtube.com
[On using Continuous Latent Space Vectors in the context windows of Transformers and LLMs] #SundayHarangue 👉 x.com/rao2z/status...
In the year since LRMs ("reasoning models") hit the scene, we have been trying to understand, analyze and demystify them.. Here are our efforts to date--conveniently all in one place..👇 www.linkedin.com/posts/subbar...
In the year since LRMs ("reasoning models") hit the scene, we have been trying to understand, analyze and demystify them.. Here are our efforts to date--conveniently all in one… | Subbarao K...
In the year since LRMs ("reasoning models") hit the scene, we have been trying to understand, analyze and demystify them.. Here are our efforts to date--conveniently all in one place.. (𝗙𝗶𝗿𝘀𝘁..) 𝗘𝘃𝗮𝗹...
linkedin.com
𝐏𝐞𝐫𝐟𝐨𝐫𝐦𝐚𝐭𝐢𝐯𝐞 𝐓𝐡𝐢𝐧𝐤𝐢𝐧𝐠? The anthropomorphization of LRM intermediate tokens as thinking begat a cottage industry to "get efficiency by shortening thinking." We ask: 𝗜𝘀 𝗖𝗼𝗧 𝗹𝗲𝗻𝗴𝘁𝗵 𝗿𝗲𝗮𝗹𝗹𝘆 𝗮 𝗿𝗲𝗳𝗹𝗲𝗰𝘁𝗶𝗼𝗻 𝗼𝗳 𝗽𝗿𝗼𝗯𝗹𝗲𝗺 𝗵𝗮𝗿𝗱𝗻𝗲𝘀𝘀 𝗼𝗿 𝗶𝘀 𝗶𝘁 𝗺𝗼𝗿𝗲 𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝘃𝗲? 👉 www.linkedin.com/posts/subbar...
Rejecting papers in #AI Conferences because of "resource constraints" is shooting ourselves in the foot as a community; use Findings.. #SundayHarangue 👇 x.com/rao2z/status...
Subbarao Kambhampati (కంభంపాటి సుబ్బారావు) on X: "Rejecting papers in #AI Conferences because of "resource constraints" is shooting ourselves in the foot as a community; use Findings.. #SundayHarangue By now, we have all know that top AI conferences are oversubscribed (in terms of paper submissions), and have heard that that" / X
Rejecting papers in #AI Conferences because of "resource constraints" is shooting ourselves in the foot as a community; use Findings.. #SundayHarangue By now, we have all know that top AI conferences are oversubscribed (in terms of paper submissions), and have heard that that
x.com
Proofs are not reasoning traces & I/O Format Language shouldn't be much of an issue for LLMs + other things #SundayHarangue (Special IMO edition). 🧵 👇 x.com/rao2z/status...
Subbarao Kambhampati (కంభంపాటి సుబ్బారావు) on X: "Proofs are not reasoning traces & I/O Format Language shouldn't be much of an issue for LLMs #SundayHarangue (Special IMO edition). 1/ My feed these last couple of days of IMO discussions has been full of comments that seem to conflate LRM intermediate tokens (aka reasoning" / X
Proofs are not reasoning traces & I/O Format Language shouldn't be much of an issue for LLMs #SundayHarangue (Special IMO edition). 1/ My feed these last couple of days of IMO discussions has been full of comments that seem to conflate LRM intermediate tokens (aka reasoning
x.com